Positron emission tomography (PET) is attracting attention as an effective test method for earlier diagnosis of cancer, cerebrovascular disease, dementia, etc. In PET, a compound labeled with a trace amount of positron emitting nuclides is administered and annihilation radiation emitted from inside the body is detected to create an image of biological functions such as sugar metabolism and check for a disease and its extent. PET devices for practicing PET have been put into actual use.
The principle of PET will be described below. Positron emitted from a positron emitting nuclide by positron decay is annihilated with ambient electron to produce a pair of 511-keV annihilation radiations, which are measured by a pair of radiation detectors based on the principle of coincidence counting. The position of the nuclide can thus be located on a single line (coincidence line) that connects the pair of detectors. An axis from the patient's head to feet will be defined as a body axis. The distribution of nuclides on a plane that perpendicularly crosses the body axis is determined by two-dimensional image reconstruction from data on coincidence lines on the plane, measured in various directions.
Early PET devices therefore have had a single-ring detector in which detectors are closely arranged in a ring shape on a plane to be the field-of-view so as to surround the field-of-view. With the advent of a multi-ring detector which includes a lot of single-ring detectors closely arranged in the direction of the body axis, the two-dimensional field-of-view has subsequently been extended to three dimensions. Since 1990's, 3D mode PET devices have been actively developed which perform coincidence measurement even between different detector rings with a significant improvement in sensitivity.
For cancer detected by the PET diagnosis or the like, treatments have a critical role. Approaches other than surgical operations and medication include radiation therapy of irradiating the affected area with radiations such as X-rays and gamma rays. In particular, particle radiotherapy of irradiating only a cancerous area with a heavy particle beam or proton beam is attracting much attention as a method both with an excellent treatment effect and a sharply concentrated irradiation characteristic with respect to the affected area. For irradiation, the direction and dose of the beam are precisely controlled according to a treatment plan which is thoroughly calculated based on X-ray CT images or the like obtained separately.
The patient positioning accuracy is the key to administer treatment in precise accordance with the treatment plan. The irradiation field is often positioned based on an X-ray image. In general, X-ray images fail to provide a sufficient contrast between tumor and normal tissue, and it is difficult to identify a tumor itself for positioning. In addition to such misalignment of the irradiation field at the time of patient setup, other problems have been pointed out such as a change in the tumor size from the time of creation of the treatment plan, and respiratory and other movements of the tumor position. Under the present circumstances, it is difficult to accurately identify whether irradiation is performed according to the treatment plan. Even if the actual irradiation field deviates from the treatment plan, it is not easy to detect.
To solve the foregoing problems, attention is being given to methods of imaging the irradiation field in real time using the PET techniques. In one study, fludeoxyglucose (FDG) or other PET medicine for use in cancer diagnosis is administered before treatment. Using a PET device installed in combination with the treatment device, imaging is performed to directly observe the tumor position while positioning the irradiation field. In another method, no PET medicine is administered. Instead, annihilation radiations occurring from a projectile fragmentation reaction, target fragmentation reaction, and/or photonuclear reaction in particle beam irradiation or X-ray irradiation are rendered into an image by using the principle of PET. Therapy monitoring is possible since the position of occurrence of the annihilation radiations has a strong correlation with the dose distribution of the irradiation beam (W. Enghardt et al., “Charged hadron tumour therapy monitoring by means of PET,” Nucl. Instrum. Methods A 525, pp. 284-288, 2004. S. Janek et al., “Development of dose delivery verification by PET imaging of photonuclear reactions following high energy photon therapy,” Phys. Med. Biol., vol. 51 (2006) pp. 5769-5783).
If a new PET medicine that allows an immediate evaluation of the activity of tumor cells and normal cells is put to actual use in the future, it will be possible to not only image the tumor position and the dose distribution but also monitor the effect of treatment on tumor cells and the impact on ambient normal cells in real time for more appropriate irradiation control.
In an ordinary PET device, detectors are arranged in a ring-like configuration. To install the detectors in combination with a treatment device, they need to be arranged so as not to block the treatment beam. Studies have so far been made on an opposed gamma camera type PET device in which two flat PET detectors are installed across the bed of the treatment device. Such a PET device has had an essential problem that the detector gap causes a lack of information necessary for image reconstruction, resulting in nonuniform resolution and lower device sensitivity (P. Crespo et al., “On the detector arrangement for in-beam PET for hadron therapy monitoring,” Phys. Med. Biol., vol. 51 (2006) pp. 2143-2163, and Satoshi Yamaguchi et al, “Development of molecular imaging device for patient setup—A basic simulation on Open-PET device—,” Japanese Journal of Medical Physics, Vol. 28, Sup. 2 (2008) pp. 256-257).
To improve the sensitivity of a PET device, as illustrated in FIG. 1(a), the detectors need to be closely arranged in a tunnel-like configuration to form a multi-ring detector 10 with an increased solid angle. The long tunnel-shaped patient port, however, increases psychological stress on the patient 6 under examination as well as obstructs the patient's treatment. In view of this, as illustrated in FIG. 1(b), the applicant has proposed an open PET device in which a plurality of multi-ring detectors (also referred to as detector rings) 11 and 12 split in the direction of the body axis of the patient 6 are arranged apart from each other to provide a physically opened area of field-of-view (also referred to as an open field-of-view). As shown in FIG. 2, images in the open area are reconstructed from the coincidence lines between the remaining multi-ring detectors 11 and 12. In the diagram, 8 represents a bed.
As shown in FIGS. 1(b) and 2, the open PET device is designed to have two split detectors of identical width (Taiga Yamaya, Taku Inaniwa, Shinichi Minohara, Eiji Yoshida, Naoko Inadama, Fumihiko Nishikido, Kengo Shibuya, Chih Fung Lam and Hideo Murayama, “A proposal of an open PET geometry,” Phy. Med. Biol., 53, pp. 757-773, 2008). The open PET device is suitable for monitoring in radiation therapy since the beam irradiation can be performed without interfering with the detectors.
As shown in FIG. 3, the field-of-view is 2W+G in the direction of the body axis, where W is the dimension (also referred to as width) of the detector rings 11 and 12 in the direction of the body axis, and G is the dimension (also referred to as gap) of the intervening open area in the direction of the body axis. As shown in FIG. 3(c), if the open area gap G exceeds W, the imaging area becomes discontinuous in the direction of the body axis. The upper limit of the open area gap G to obtain a field-of-view continuous in the direction of the body axis is thus W as shown in FIG. 3(b). The sensitivity concentrates at the center of the open area and drops significantly in the periphery of the open area. To suppress the extreme sensitivity drops at both ends of the open area, G needs to be set smaller than W as shown in FIG. 3(a). This, however, narrows the open area gap and the field-of-view in the direction of the body axis (see the foregoing document).
Since the open PET device previously proposed by the applicant has had the problem that the sensitivity concentrates at the center of the open area and drops significantly in the periphery of the open area, it has been needed to increase W relative to G in order to suppress the local sensitivity drops. Since the open area gap is limited by W, it has been needed to increase W itself in order to increase the open area gap further. However, there has been the problem that the increased number of detectors to constitute each multi-ring detector makes the device higher in price, larger in size, and more complicated in configuration.
In particular, when the open PET device is used to perform monitoring in particle radiotherapy, the beam irradiation may cause a performance drop or a failure of the detectors in cases such as when the detector circuit itself is affected. During the beam irradiation, the detectors therefore need to be separated from the irradiation field by several tens of centimeters. In order to expand the open area gap G, as mentioned previously, the dimension W of the detectors in the direction of the body axis need to be increased. Such an increase undesirably results in higher price, larger size, and further complication of the device. The detectors may alternatively be moved to near the irradiation field for PET measurement after the end of the irradiation. This, however, has the problem that the patient's binding time increases as much as the moving time. In addition, nuclides produced by the beam irradiation have an extremely short half-life of about several tens of seconds to 20 minutes, and the nuclides can even move within the living body due to the blood flow and other factors. It is therefore desirable to start PET measurement immediately after the irradiation.